Benchgen

MultiChallenge — Results

RankModelScore
1raon-1-mini-preview59
2nova-2-pro0.777
3nova-2-lite0.766
4nova-2-omni0.755
5gpt-50.696
6qwen3-5-397b-a17b0.676
7nemotron-3-ultra-550b-a55b0.638
8qwen3-5-122b-a10b0.615
9qwen3-5-27b0.608
10o30.604
11qwen3-5-35b-a3b0.6
12nemotron-3-super-120b-a12b0.552
13qwen3-5-9b0.545
14kimi-k2-instruct-09050.541
15kimi-k2-instruct0.541
16mai-thinking-10.53
17qwen3-5-4b0.49
18minimax-m1-40k0.447
19minimax-m1-80k0.447
20gpt-4-50.438
21o4-mini0.43
22gpt-4o0.403
23o3-mini0.399
24nemotron-3-nano-30b-a3b0.385
25gpt-4-10.383

MultiChallenge — Phases

1 phaseActive

Phase 1·

MultiChallenge

Aug 3, 2026
Current